基于混合卷积窗和双窗法的FFT算法研究

Research on FFT Based on Xixed ConvolutionWindow and DoubleWindow Method

FFT算法是信号处理中一个不可或缺的部分,也是其中需要改进的部分。设计一个精度优良的FFT算法有助于推进频谱分析的实用化进程。针对FFT改进算法的实现需求,文章采用了C语言结构设计了一个任意点数的FFT算法,分析了混合卷积窗的频谱特性,并总结了任意窗函数的幅值恢复方法。最终通过构建混合卷积窗和双窗法结合的处理方法有效提高了FFT算法的精度,仿真结果表明,与普通FFT、混合卷积窗、双窗法方法相比,基于混合卷积窗和双窗法结合的FFT算法具有更高的精度。

FFT is an indispensable part of signal processing,and also needs to be improved.Design-ing a good FFT algorithm is helpful to advance the practical process of spectrum analysis.To meet the requirement of FFT improved algorithm,this paper uses C language structure to design an FFT with arbitrary points,analyzes the spectrum characteristics of mixed convolution window,and sum-marizes the amplitude recovery method of arbitrary window function.Finally,the FFT algorithm is effectively improved by constructing the combination of mixed convolution window and double-window method.The simulation results show that the FFT algorithm based on the combination of mixed convolution window and double-window method has higher accuracy than the common FFT,mixed convolution window and double-window method.